2020-21 Donruss

• G-001 cardboardconnection.com/2020-21-donruss-basketball-nba-cards (the spec's primary source) could not be read: WebFetch returns ROBOTS_DISALLOWED and direct HTTPS is blocked by the session egress proxy (403 on CONNECT). No cardboardconnection data was used for this release.
• G-002 beckett.com/news/2020-21-donruss-basketball-cards/ returned HTTP 403 and could not be used as a cross-check for base set size, parallels, or insert/auto counts.
• G-003 Choice Black base parallel is listed by sportscardchecklist (250 cards) but no source consulted documents its print run or whether it is serial numbered. Row OMITTED from parallels.csv rather than guessing; parallel count is therefore 21, not 22.
• G-004 Print runs for Press Proof Silver (/349), Holo Yellow Laser (/25) and Holo Gold Laser (/10) are single-source (checklistcenter only); not independently confirmed.
• G-005 Channel is undocumented for 11 of 21 base parallels (Press Proof Silver, Press Proof Purple, Holo Red Laser, Holo Blue Laser, Holo Yellow Laser, Holo Red and Blue Laser, Holo Gold Laser, Holo Green Laser, Holo Light Blue Laser, Holo Red and Gold Laser, Yellow Flood); recorded as UNKNOWN. No source consulted maps these to hobby/blaster/hanger/cello. [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.]
• G-006 Holo Red and Gold Laser is reported as a Tmall (Asia) exclusive only by a break-video title; no checklist or retailer source confirms it, so channel is recorded as UNKNOWN. [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.]
• G-007 Parallel naming varies by source: sportscardchecklist and Panini retail copy print 'Holo <colour> Laser'; checklistcenter prints the short form ('Purple Laser', 'Red & Blue Laser'). The 'Holo ...' long form was recorded. Ampersand vs 'and' also varies.
• G-008 Holo Orange Laser and Holo Purple Laser channel is inferred from dacardworld blaster product copy that names them as findable in blasters; the copy does NOT state they are blaster-exclusive. Channel 'blaster' may be too narrow. [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.]
• G-009 Dominator Signatures and Hall Dominator Signatures card-count disagreement: checklistcenter lists 30 each; sportscardchecklist lists 29 each (its Black/Gold parallels list 30). Recorded 30. Set-name spelling also differs (Dominator vs Dominators).
• G-010 Rookie Jersey Kings Prime (40) is single-source (sportscardchecklist only); checklistcenter lists Jersey Kings Prime and Jersey Series Prime but not this one.
• G-011 Prime memorabilia parallels (Jersey Kings Prime, Rookie Jersey Kings Prime, Jersey Series Prime) are recorded as separate insert-type sets because both checklist sources enumerate them separately; they are arguably #/10 parallels of their parent relic sets.
• G-012 Insert and autograph card-level checklists were not captured (out of scope per spec); only set names and counts are recorded, so per-card verification of those counts was not possible.
• G-013 Base cards 226-250 were enumerated in full only by breakninja; checklistcenter's rendering truncated after card 225 (it confirmed the 226-250 range contains James Wiseman, Patrick Williams, Onyeka Okongwu ... Grant Riller, consistent with breakninja's order).
• G-014 rookie_flag is set to 'no' for 1-200 and 'yes' for 201-250 based on the sets' structural description (200 veterans + 50 Rated Rookies) and secondary-market catalogs tagging 201-250 as (RC); no consulted checklist prints explicit per-card RC designations. Note #212 Caleb Martin debuted in 2019-20, so 'Rated Rookie' here does not strictly equal RC.
• G-015 Base set size phrasing differs: checklistcenter reports 'Base Set 200' plus 'Rated Rookies Base 50'; sportscardchecklist and tcdb report a single 250-card set. Recorded 250 with Rated Rookies inside the base numbering, per the brand rule for paper Donruss.
• G-016 The Choice parallel family (Choice, Choice Red, Choice Blue, Choice Gold, Choice Black Gold) comes from the separately sold 2020-21 Donruss Choice retail box, which shares this checklist. Included here as base parallels with channel 'choice'; if the ontology treats Donruss Choice as its own release these rows should move. [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-17: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: still open; 11 of 21 parallels lack a sourced channel.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.] [REVIEWED 2026-08-19: count is now 67 of 77 - the channel pass resolved -56 of these with cited evidence. Gap remains open for the rest.]
• G-017 Choice Gold (/10) and Choice Black Gold (/8) print runs come from price-guide and marketplace catalog pages rather than a checklist source; not confirmed by checklistcenter or sportscardchecklist.
• G-018 Retail street date (2021-03-10, dacardworld blaster page) differs from the hobby release date (2021-03-03). release.csv records the hobby date per spec.
• G-019 Source-calibration record for this release's insert/autograph card numbers. Calibrated on a CODED set, not on base cards: for Signature Series, tcdb.com prints letter codes (SG-PAC, SG-ZWL, ...) and eBay listing titles confirm them ("#SG-JPT", "#SG-SBE", "#SG-TSN", "#SG-DBG", "#SG-SWT", "#SG-JRA", "#SG-BSC", "#SG-ACL"), while breakninja.com AND sportscardchecklist.com both renumber the same set 1..100 in a completely different order (their #1 is Zion Williamson). Those two sites were therefore used only for player rosters and for corroborating the plainly-numbered inserts, never for card_number on a coded set. Every autograph and memorabilia set in this release is coded (SG-, CS-, DS-, HD-, JK-, RJ-, JS-); the twelve plain inserts genuinely are 1..N and breakninja agrees with tcdb card-for-card on all twelve. The one exception in both directions is Rated Rookies Signatures, which is an autograph set but is NOT coded: it reuses the Rated Rookies base numbers 201-250 (confirmed by sportscardinvestor.com "#201" and an eBay title "LaMelo Ball #202 Signatures (AU, RC)"), which is why its rows legitimately mirror base_cards.csv 201-250.
• G-020 Duplicate printed card code inside one set, recorded as tcdb prints it and NOT disambiguated by us: "SG-JWS" is listed for BOTH Jerry West and James Wiseman. tcdb shows the pair in all three of its sort orders (card number, last name, first name), and sportscardspro.com's disambiguating slug 'james-wiseman-sg-jws' independently gives Wiseman SG-JWS; comc.com catalogues the sibling Optic card as "#SS-JWS.1 - James Wiseman", the '.1' suffix being how COMC handles a genuinely reused Panini code, so the collision may be real rather than a tcdb error. No marketplace listing was found printing a different code for either player. Both rows are written verbatim; this VIOLATES the chase_cards UNIQUE (set_id, card_number) constraint and needs a human decision before load.
• G-021 Duplicate printed card code inside one set, recorded as tcdb prints it and NOT disambiguated by us: "CS-JRS" is listed for BOTH Jason Richardson and Jalen Rose, on tcdb's card-number and first-name sort orders alike. The code fits either player under Panini's initial-plus-two-surname-letters convention (RichardSon / RoSe). An eBay title for the sibling Optic set gives Jason Richardson "#SS-JRI", which hints that one of the two paper codes differs, but no listing printing a CS- code for either player was found. Both rows are written verbatim; this VIOLATES the chase_cards UNIQUE (set_id, card_number) constraint and needs a human decision before load.
• G-022 PARTIAL: 28 of the 30 cards in sets.csv were written. (a) Ralph Sampson (Houston Rockets) appears on breakninja.com's 30-player list for this set but is absent from tcdb entirely, and tcdb is the only source consulted that prints the HD- codes; no code could be sourced, so the row was omitted rather than guessed. (b) tcdb prints "HD-DRJ" for BOTH Dino Radja and Elvin Hayes. HD-DRJ fits Radja under Panini's initial-plus-two-surname-letters convention and does not fit Hayes at all (which would give HD-EHY), so the Elvin Hayes row was omitted rather than written with a code that is demonstrably Radja's. Searches for HD-EHY / HD-ELH / HD-EHS returned nothing for this product. Elvin Hayes IS in the set per breakninja.
• G-023 PARTIAL: 29 of the 30 cards in sets.csv were written. DeAndre Jordan (Brooklyn Nets) is on breakninja.com's 30-player list but absent from tcdb (whose set header states "Total Cards: 29"), and tcdb is the only source consulted that prints the DS- codes; no DS- code for him could be sourced, so the row was omitted rather than guessed.
• G-024 Resolves the count half of G-009 in favour of 30. breakninja.com lists 30 distinct players for BOTH Dominator Signatures (the 30th being DeAndre Jordan) and Hall Dominator Signatures (the 30th being Ralph Sampson), matching checklistcenter's 30 and contradicting sportscardchecklist's 29. tcdb also holds 29 for each and is missing exactly those two players, so tcdb's and sportscardchecklist's 29 look like the same single omission rather than independent confirmation. sets.csv's card_count of 30 is left unchanged and is believed correct; the naming half of G-009 (Dominator vs Dominators) is untouched.
• G-025 Code disagreement for LaMelo Ball. tcdb prints "SG-LMB" and is internally consistent about it across three different sort orders (in card-number order it falls between SG-LGW and SG-LNJ). sportscardspro.com prints "#SG-LAM" for the same card (page: https://www.sportscardspro.com/game/basketball-cards-2020-donruss-signature-series/lamelo-ball-sg-lam). No eBay or COMC listing for the paper Donruss card was found to break the tie; searches kept returning the Donruss Optic version. Recorded SG-LMB (tcdb). NOTE: the sibling Rookie Jersey Kings set uses RJ-LAM for LaMelo, so SG-LAM is not implausible.
• G-026 Sourcing method note. tcdb's default checklist view truncates partway through long sets, so Signature Series (98) and Jersey Series (100) were assembled by unioning tcdb's card-number, last-name and age sort orders for the same set id, and the resulting rosters were then checked against breakninja's independent 100/98-player lists to confirm nothing was missed. One row - Zion Williamson in Jersey Series - was reachable in none of the three tcdb views; its code JS-ZWL comes from the eBay listing title "2020-21 Panini Donruss Jersey Series Zion Williamson #JS-ZWL". Teams for six Signature Series rows (SG-IST, SG-OBI, SG-TTE, SG-TYG, SG-XTL, SG-ZWL) came from the same player's row on another 2020-21 Donruss tcdb set page because tcdb's age-sorted view prints no team column.
• G-027 Set-name and count observations against outside sources, recorded rather than acted on. (a) tcdb spells set 'Jersey Kings Prime' as "Jerseys Kings Prime" (sid 262365); our set_name is unchanged. (b) breakninja and sportscardchecklist both render Signature Series as 100 rows, while tcdb and sets.csv agree on 98 - the two aggregators are renumbering this coded set and their row count is not evidence about set size. (c) Rookie Jersey Kings Prime, recorded in G-010 as single-source (sportscardchecklist only), now has a second independent source: tcdb sid 262397 lists it with 40 cards carrying the same RJ- codes as the base Rookie Jersey Kings run.
• G-028 cardboardconnection.com remained unreadable for this pass as well (WebFetch ROBOTS_DISALLOWED / egress proxy 403 on direct HTTPS), as did beckett.com (403 on both /news/ and /basketball/ set pages) and comc.com set pages (403), though COMC card URLs surfaced through web search were usable as corroboration. No cardboardconnection or beckett data was used for the insert/autograph checklists. This extends G-001 and G-002 to the checklist pass.
• G-029 Player-name spelling disagreement with existing data, recorded not corrected. tcdb prints "Jahmi'us Ramsey" for card 247 and that is what insert_cards.csv carries (names are written as the source prints them); base_cards.csv carries "Jahmius Ramsey" for the same card number 247. This is the only one of the 50 Rated Rookies Signatures rows that does not match base_cards.csv on both card_number and player_name. base_cards.csv was not touched.
• G-030 RETROACTIVE TRUNCATION-RECOVERY SWEEP 2026-09-07 (Gallery.cfm?PageIndex=N, Errors.cfm). TRUNCATION RULED OUT for both short sets; only 1 row was recoverable. Dominator Signatures: Gallery states Total Cards 29 and renders exactly 29 across 3 pages, matching our 29 held rows card-for-card; sets.csv says 30, so the 30th card is absent from tcdb entirely, NOT hidden by truncation, and no second source was found for it. Hall Dominator Signatures: Gallery states Total Cards 29 against our 28; the one extra entry is recovered (HD-DRJ Elvin Hayes, see separate gap), leaving 29 of the 30 sets.csv claims - again a genuine tcdb shortfall, not truncation. Errors.cfm returned no entries. team_or_school left empty on the recovered row: Gallery renders no team.
• G-031 AMBIGUOUS DUPLICATE NEEDING REVIEW - 2020-21 Hall Dominator Signatures HD-DRJ. tcdb assigns HD-DRJ to TWO players, Dino Radja (already held) and Elvin Hayes (appended 2026-09-07), confirmed by two independent tcdb endpoints (Gallery.cfm page 1 and Errors.cfm, the latter naming 'both Elvin Hayes (HD-DRJ) and Dino Radja (HD-DRJ)'). It is UNRESOLVED whether this is a genuine initials collision or a tcdb copy-down error from the immediately preceding Radja row: the cid-adjacency test that normally discriminates a smear from a collision at n=2 CANNOT FIRE here, because this page's ViewCard cids run strictly sequentially in display order (16750919..16750928), which the spec states disarms the test. Per the spec the ambiguous duplicate is therefore WRITTEN AND FLAGGED rather than guessed. It was deliberately NOT 'corrected' to a code fitting Hayes's initials: the spec is explicit that Panini codes are not derivable from initials. A marketplace listing title showing the printed number on an Elvin Hayes Hall Dominator Signatures card would settle it.

If you know a source that answers one of these, tell us — name the question's label and where to look.